| 123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269 | /* boost random/shuffle_order.hpp header file * * Copyright Jens Maurer 2000-2001 * Copyright Steven Watanabe 2010 * Distributed under the Boost Software License, Version 1.0. (See * accompanying file LICENSE_1_0.txt or copy at * http://www.boost.org/LICENSE_1_0.txt) * * See http://www.boost.org for most recent version including documentation. * * $Id$ * */#ifndef BOOST_RANDOM_SHUFFLE_ORDER_HPP#define BOOST_RANDOM_SHUFFLE_ORDER_HPP#include <iostream>#include <algorithm>     // std::copy#include <cassert>#include <boost/config.hpp>#include <boost/limits.hpp>#include <boost/static_assert.hpp>#include <boost/cstdint.hpp>#include <boost/random/detail/operators.hpp>#include <boost/random/detail/seed.hpp>#include <boost/random/detail/signed_unsigned_tools.hpp>#include <boost/random/linear_congruential.hpp>#include <boost/random/detail/disable_warnings.hpp>namespace boost {namespace random {/** * Instatiations of class template @c shuffle_order_engine model a * \pseudo_random_number_generator. It mixes the output * of some (usually \linear_congruential_engine) * \uniform_random_number_generator to get better statistical properties. * The algorithm is described in * *  @blockquote *  "Improving a poor random number generator", Carter Bays *  and S.D. Durham, ACM Transactions on Mathematical Software, *  Vol 2, No. 1, March 1976, pp. 59-64. *  http://doi.acm.org/10.1145/355666.355670 *  @endblockquote * * The output of the base generator is buffered in an array of * length k. Every output X(n) has a second role: It gives an * index into the array where X(n+1) will be retrieved. Used * array elements are replaced with fresh output from the base * generator. * * Template parameters are the base generator and the array * length k, which should be around 100. */template<class UniformRandomNumberGenerator, std::size_t k>class shuffle_order_engine{public:    typedef UniformRandomNumberGenerator base_type;    typedef typename base_type::result_type result_type;    BOOST_STATIC_CONSTANT(bool, has_fixed_range = false);    BOOST_STATIC_CONSTANT(std::size_t, buffer_size = k);    BOOST_STATIC_CONSTANT(std::size_t, table_size = k);    BOOST_STATIC_ASSERT(std::numeric_limits<result_type>::is_integer);    BOOST_STATIC_ASSERT(k > 0);    /**     * Constructs a @c shuffle_order_engine by invoking the     * default constructor of the base generator.     *     * Complexity: Exactly k+1 invocations of the base generator.     */    shuffle_order_engine() : _rng() { init(); }    /**     * Constructs a @c shuffle_output_engine by invoking the one-argument     * constructor of the base generator with the parameter seed.     *     * Complexity: Exactly k+1 invocations of the base generator.     */    BOOST_RANDOM_DETAIL_ARITHMETIC_CONSTRUCTOR(shuffle_order_engine,                                               result_type, s)    { _rng.seed(s); init(); }    BOOST_RANDOM_DETAIL_SEED_SEQ_CONSTRUCTOR(shuffle_order_engine, SeedSeq, seq)    { _rng.seed(seq); init(); }    /**     * Constructs a @c shuffle_output_engine by using a copy     * of the provided generator.     *     * Precondition: The template argument UniformRandomNumberGenerator     * shall denote a CopyConstructible type.     *     * Complexity: Exactly k+1 invocations of the base generator.     */    explicit shuffle_order_engine(const base_type & rng) : _rng(rng) { init(); }#ifndef BOOST_NO_CXX11_RVALUE_REFERENCES    explicit shuffle_order_engine(base_type&& rng) : _rng(rng) { init(); }#endif    template<class It> shuffle_order_engine(It& first, It last)      : _rng(first, last) { init(); }    void seed() { _rng.seed(); init(); }    /**     * Invokes the one-argument seed method of the base generator     * with the parameter seed and re-initializes the internal buffer array.     *     * Complexity: Exactly k+1 invocations of the base generator.     */    BOOST_RANDOM_DETAIL_ARITHMETIC_SEED(shuffle_order_engine,        result_type, seed_arg)    { _rng.seed(seed_arg); init(); }    /**     * Invokes the one-argument seed method of the base generator     * with the parameter seq and re-initializes the internal buffer array.     *     * Complexity: Exactly k+1 invocations of the base generator.     */    BOOST_RANDOM_DETAIL_SEED_SEQ_SEED(shuffle_order_engine, SeedSeq, seq)    { _rng.seed(seq); init(); }    template<class It> void seed(It& first, It last)    { _rng.seed(first, last); init(); }    const base_type& base() const { return _rng; }    result_type operator()() {        // calculating the range every time may seem wasteful.  However, this        // makes the information locally available for the optimizer.        typedef typename boost::random::traits::make_unsigned<result_type>::type base_unsigned;        const base_unsigned brange =            detail::subtract<result_type>()((max)(), (min)());        const base_unsigned off =            detail::subtract<result_type>()(y, (min)());        base_unsigned j;        if(k == 1) {            j = 0;        } else if(brange < (std::numeric_limits<base_unsigned>::max)() / k) {            // try to do it in the native type if we know that it won't            // overflow            j = k * off / (brange + 1);        } else if(brange < (std::numeric_limits<uintmax_t>::max)() / k) {            // Otherwise try to use uint64_t            j = static_cast<base_unsigned>(                static_cast<uintmax_t>(off) * k /                (static_cast<uintmax_t>(brange) + 1));        } else {            boost::uintmax_t divisor =                static_cast<boost::uintmax_t>(brange) + 1;            j = static_cast<base_unsigned>(detail::muldiv(off, k, divisor));        }        // assert(0 <= j && j < k);        y = v[j];        v[j] = _rng();        return y;    }    /** Advances the generator by z steps. */    void discard(boost::uintmax_t z)    {        for(boost::uintmax_t j = 0; j < z; ++j) {            (*this)();        }    }    /** Fills a range with pseudo-random values. */    template<class Iter>    void generate(Iter first, Iter last)    { detail::generate_from_int(*this, first, last); }    /** Returns the smallest value that the generator can produce. */    static result_type min BOOST_PREVENT_MACRO_SUBSTITUTION ()    { return (base_type::min)(); }    /** Returns the largest value that the generator can produce. */    static result_type max BOOST_PREVENT_MACRO_SUBSTITUTION ()    { return (base_type::max)(); }    /** Writes a @c shuffle_order_engine to a @c std::ostream. */    BOOST_RANDOM_DETAIL_OSTREAM_OPERATOR(os, shuffle_order_engine, s)    {        os << s._rng;        for(std::size_t i = 0; i < k; ++i)            os << ' ' << s.v[i];        os << ' ' << s.y;        return os;    }    /** Reads a @c shuffle_order_engine from a @c std::istream. */    BOOST_RANDOM_DETAIL_ISTREAM_OPERATOR(is, shuffle_order_engine, s)    {        is >> s._rng;        for(std::size_t i = 0; i < k; ++i)            is >> std::ws >> s.v[i];        is >> std::ws >> s.y;        return is;    }    /** Returns true if the two generators will produce identical sequences. */    BOOST_RANDOM_DETAIL_EQUALITY_OPERATOR(shuffle_order_engine, x, y)    { return x._rng == y._rng && x.y == y.y && std::equal(x.v, x.v+k, y.v); }    /** Returns true if the two generators will produce different sequences. */    BOOST_RANDOM_DETAIL_INEQUALITY_OPERATOR(shuffle_order_engine)private:    /// \cond show_private    void init()    {        // we cannot use std::generate, because it uses pass-by-value for _rng        for(result_type * p = v; p != v+k; ++p)            *p = _rng();        y = _rng();    }    /// \endcond    base_type _rng;    result_type v[k];    result_type y;};#ifndef BOOST_NO_INCLASS_MEMBER_INITIALIZATION//  A definition is required even for integral static constantstemplate<class URNG, std::size_t k>const bool shuffle_order_engine<URNG, k>::has_fixed_range;template<class URNG, std::size_t k>const std::size_t shuffle_order_engine<URNG, k>::table_size;template<class URNG, std::size_t k>const std::size_t shuffle_order_engine<URNG, k>::buffer_size;#endif/** * According to Harry Erwin (private e-mail), the specialization * @c kreutzer1986 was suggested in: * * @blockquote * "System Simulation: Programming Styles and Languages (International * Computer Science Series)", Wolfgang Kreutzer, Addison-Wesley, December 1986. * @endblockquote */typedef shuffle_order_engine<    linear_congruential_engine<uint32_t, 1366, 150889, 714025>,    97> kreutzer1986;/** * The specialization @c knuth_b is specified by the C++ standard. * It is described in * * @blockquote * "The Art of Computer Programming, Second Edition, Volume 2, * Seminumerical Algorithms", Donald Knuth, Addison-Wesley, 1981. * @endblockquote */typedef shuffle_order_engine<minstd_rand0, 256> knuth_b;} // namespace randomusing random::kreutzer1986;} // namespace boost#include <boost/random/detail/enable_warnings.hpp>#endif // BOOST_RANDOM_SHUFFLE_OUTPUT_HPP
 |